positional shift in the action of the robot in the specified Frame

    Guys;

    We are using a Fanuc Robot m-710ic/70 with R30ia controller mechanism. We are making deburring process with this system. It is observed that; at our implemented parts we are getting almost .1'' of an offset, while they were working properly a month ago. Luckily we were using soft tools with the programs and no big harm occured.

    The real issue here is; we are not getting the offset just in one direction, at some of my processes it is working above the taught points, at some below of them, and some of the programs shifted in the left and the right. So there doesn't seem to be a problem with the rotary.

    Another important thing about the problem is, we don't encounter this problem if we don't begin the part processes; what I mean is there is no problem with the Tool picking or releasing while we are using another frame. But when I begin my part processes, that is when we shift to another frame, we are getting this offset.

    Now I am going to do a calibration procedure to see if there is any shift in the frames.

    My questions are;
    What might be the cause of this problem? I am for sure that, there hasn't been any crash in the cell..

    And what should I do next? I will be posting the calibration results as I get them. I will let you know if there is any change or not.

    I will appreciate anything from you guys. Thanks from now.

    (Un?)fortunately, the frame results look the same. Do you guys have any idea of the cause of the problem or how to fix it?

    I will appreciate any help.

    If your toll pick/drop is fine and I mean excatly like it was a month ago then I can assume is not the robot location or tooling
    If your user frames are (again) exactly like before, which BTW, explain how do you use the frame ? do you touch it up everytime ?

    Then , just like easyrobot said "fixture are being mislocated" or part misplaced in fixture
    Could you write a 3 point program (don use user frames) to "mark" where your fixture is ?

    Retired but still helping

    I know everything looks like a fixturing or a part locating problem; in addition to that, our cell, has an integrated vision locating system, which I had disabled and used pinning instead, for locating the part. Also the frame I am having problem is located on the rotary system.

    But unfortunately, I am not having this positional shift in one direction. What I mean is, in one item the end effector is now going to a point much more lower compared to the taught position and in one it is going to a higher point. In one, it is going to a point more located at left and vice versa. :hmmm: and we encountered this problem at all of the parts.

    Easyrobot; sorry I didn't mean that we are having offset results on frames, instead we are using the same uframe each time. The offset is at the position of the end effector which I observed by eye.

    Fabian Munoz;
    I am not one hundred percent sure about the tool picking is just like the way it was a month ago. but as I can see if it was shifted, as my process programs shifted, then there would be an absolute crash, now they seem to be working properly.

    And another thing is, my frames are not exactly like before. It is off about 0.1 mm in each axis and 0.1 degrees in each angles, which I ignored.

    For locating the part on the frame, we have a vision system. I touch up the points on the uFrame when I start working on a new part, once I get good results I freeze the process parameters and process points.

    Edited once, last by omfidan (February 11, 2013 at 7:54 AM).

    Well,
    From this point of view, it looks like an accumulative error.
    Something to consider is the rotating system. it might not be leveled correctly or inclusive it might not be rotating on a perfect circle, (elliptical) this kind of error explain why in some points you have the error in one direction and in other points the error is in another direction. Something is getting loose .... Maybe.

    Are all the wrong points related to this rotating system? Or the wrong points are all over the cell??

    I have to ask, did you double-triple check your user frame is the same as when you program the part first time??

    I would recommend you to program some routines to check the integrity of the mechanical system, as suggested before, have a program of 3 point to check some marks on the fixture always related to Uframe 0

    Guys;

    Thank you for your responses, we couldn't find the cause nor a solution. So I touched up the points and now it is working properly.

    But at least we can check if there is a mismatch before we can start an operation. But I don't see why we should test the fixture with 3 points, in frame 0 instead of the frame attached to the rotary?

Advertising from our partners